Search Results (773)

The rapid development of deepfake technologies has led to the widespread proliferation of facial image forgeries, raising significant concerns over identity theft and the spread of misinformation. Although recent dual-domain detection approaches that integrate spatial and frequency features have achieved noticeable progress, they still suffer from limited sensitivity to local forgery regions and inadequate interaction between spatial and frequency information in practical applications. To address these challenges, we propose a novel forgery-aware guided spatial–frequency feature fusion network. A lightweight U-Net is employed to generate pixel-level saliency maps by leveraging structural symmetry and semantic consistency, without relying on ground-truth masks. These maps dynamically guide the fusion of spatial features (from an improved Swin Transformer) and frequency features (via Haar wavelet transforms). Cross-domain attention, channel recalibration, and spatial gating are introduced to enhance feature complementarity and regional discrimination. Extensive experiments conducted on two benchmark face forgery datasets, FaceForensics++ and Celeb-DFv2, show that the proposed method consistently outperforms existing state-of-the-art techniques in terms of detection accuracy and generalization capability. The future work includes improving robustness under compression, incorporating temporal cues, extending to multimodal scenarios, and evaluating model efficiency for real-world deployment. Full article

When monitoring seed positions in soil using ultrasonic waves, the main challenge is obtaining acoustic wave characteristics at the seed locations. This study developed a three-dimensional ultrasonic model with the double media of seed–soil using the discrete element method to visualize signal variations and analyze propagation characteristics. The effects of the compression ratio (0/6/12%), excitation frequency (20/40/60 kHz), and amplitude (5/10/15 μm) on signal variation and attenuation were analyzed. The results show consistent trends: time/frequency domain signal intensity increased with a higher compression ratio and amplitude but decreased with frequency. Comparing ultrasonic signals at soil particles before and after the seed along the propagation path shows that the seed significantly absorbs and attenuates ultrasonic waves. Time domain intensity drops 93.99%, and first and residual wave frequency peaks decrease by 88.06% and 96.39%, respectively. Additionally, comparing ultrasonic propagation velocities in the double media of seed–soil and the single soil medium reveals that the velocity in the seed is significantly higher than that in the soil. At compression ratios of 0%, 6%, and 12%, the sound velocity in the seed is 990.47%, 562.72%, and 431.34% of that in the soil, respectively. These findings help distinguish seed presence and provide a basis for ultrasonic seed position monitoring after sowing. Full article

In neural video compression, an approximation of the target frame is predicted, and a mask is subsequently applied to it. Then, the masked predicted frame is subtracted from the target frame and fed into the encoder along with the conditional information. However, this structure has two limitations. First, in the pixel domain, even if the mask is perfectly predicted, the residuals cannot be significantly reduced. Second, reconstructed features with abundant temporal context information cannot be used as references for compressing the next frame. To address these problems, we propose Conditional Masked Feature Residual (CMFR) Coding. We extract features from the target frame and the predicted features using neural networks. Then, we predict the mask and subtract the masked predicted features from the target features. Thereafter, the difference is fed into the encoder with the conditional information. Moreover, to more effectively remove conditional information from the target frame, we introduce a Scaled Feature Fusion (SFF) module. In addition, we introduce a Motion Refiner to enhance the quality of the decoded optical flow. Experimental results show that our model achieves an 11.76% bit saving over the model without the proposed methods, averaged over all HEVC test sequences, demonstrating the effectiveness of the proposed methods. Full article

In this paper, we present a new space-time Galerkin-like method, where we treat the discretization of spatial and temporal domains simultaneously. This method utilizes a mixed formulation of the tensor-train (TT) and quantized tensor-train (QTT) (please see Section Tensor-Train Decomposition), designed for the finite element discretization (Q1-FEM) of the time-dependent convection–diffusion–reaction (CDR) equation. We reformulate the assembly process of the finite element discretized CDR to enhance its compatibility with tensor operations and introduce a low-rank tensor structure for the finite element operators. Recognizing the banded structure inherent in the finite element framework’s discrete operators, we further exploit the QTT format of the CDR to achieve greater speed and compression. Additionally, we present a comprehensive approach for integrating variable coefficients of CDR into the global discrete operators within the TT/QTT framework. The effectiveness of the proposed method, in terms of memory efficiency and computational complexity, is demonstrated through a series of numerical experiments, including a semi-linear example. Full article

Driven by dual carbon targets, the scale of wind power integration has surged dramatically. However, its strong volatility causes insufficient short-term prediction accuracy, severely constraining grid security and economic dispatch. To address three key challenges in extracting temporal characteristics of strong volatility, adaptive fusion of multi-source features, and enhancing model interpretability, this paper proposes a Time-Domain Dual-Channel Adaptive Learning Model (TDDCALM). The model employs dual-channel feature decoupling: one Transformer encoder layer captures global dependencies while the raw state layer preserves local temporal features. After TCN-based feature compression, an adaptive weighted early fusion mechanism dynamically optimizes channel weights. The ACON adaptive activation function autonomously learns optimal activation patterns, with fused features visualized through visualization techniques. Validation on two wind farm datasets (A/B) demonstrates that the proposed method reduces RMSE by at least 8.89% compared to the best deep learning baseline, exhibits low sensitivity to time window sizes, and establishes a novel paradigm for forecasting highly volatile renewable energy power. Full article

With the rapid development of video encoding technology in the field of computer vision, the demand for tasks such as video frame reconstruction, denoising, and super-resolution has been continuously increasing. However, traditional video encoding methods typically focus on extracting spatial or temporal domain information, often facing challenges of insufficient accuracy and information loss when reconstructing high-frequency details, edges, and textures of images. To address this issue, this paper proposes an innovative LadderConv framework, which combines discrete wavelet transform (DWT) with spatial and channel attention mechanisms. By progressively recovering wavelet subbands, it effectively enhances the video frame encoding quality. Specifically, the LadderConv framework adopts a stepwise recovery approach for wavelet subbands, first processing high-frequency detail subbands with relatively less information, then enhancing the interaction between these subbands, and ultimately synthesizing a high-quality reconstructed image through inverse wavelet transform. Moreover, the framework introduces spatial and channel attention mechanisms, which further strengthen the focus on key regions and channel features, leading to notable improvements in detail restoration and image reconstruction accuracy. To optimize the performance of the LadderConv framework, particularly in detail recovery and high-frequency information extraction tasks, this paper designs an innovative ResGroup module. By using multi-layer convolution operations along with feature map compression and recovery, the ResGroup module enhances the network’s expressive capability and effectively reduces computational complexity. The ResGroup module captures multi-level features from low level to high level and retains rich feature information through residual connections, thus improving the overall reconstruction performance of the model. In experiments, the combination of the LadderConv framework and the ResGroup module demonstrates superior performance in video frame reconstruction tasks, particularly in recovering high-frequency information, image clarity, and detail representation. Full article

Extensive numerical simulations were performed in MATLAB R2020b based on the classical nonlinear Thomson scattering theory and single-electron model, to systematically examine the influence of initial phase in tightly focused linearly polarized laser pulses on the radiation characteristics of multi-energy-level electrons. Through our research, we have found that phase variation from 0 to 2π induces an angular bifurcation of peak radiation intensity, generating polarization-aligned symmetric lobes with azimuthal invariance. Furthermore, the bimodal polar angle decreases with the increase of the initial energy. This phase-controllable bimodal distribution provides a new solution for far-field beam shaping. Significantly, high-harmonic intensity demonstrates π-periodic phase-dependent modulation. Meanwhile, the time-domain pulse width also exhibits 2π-cycle modulation, which is synchronized with the laser electric field period. Notably, electron energy increase enhances laser pulse peak intensity while compressing its duration. The above findings demonstrate that the precise control of the driving laser’s initial phase enables effective manipulation of the radiation’s spatial characteristics. Full article

We demonstrate a novel bidirectional mode-locked ultrafast fiber laser based on an asymmetric dual-cavity architecture that enables freely tunable repetition rate differentials at the kilohertz level, while maintaining inherent common-mode noise suppression through precision thermomechanical stabilization. Through cascaded amplification and nonlinear temporal compression, we obtained bidirectional pulse durations of 33.2 fs (clockwise) and 61.6 fs (counterclockwise), respectively. The developed source demonstrates exceptional capability for asynchronous optical sampling applications, particularly in enabling the compact implementation of real-time measurement systems such as terahertz time-domain spectroscopy (THz-TDS) systems. Full article

Recent advancements in deep learning have driven the rapid proliferation of deepfake generation techniques, raising substantial concerns over digital security and trustworthiness. Most current detection methods primarily focus on spatial or frequency domain features but show limited effectiveness when dealing with compressed videos and cross-dataset scenarios. Observing that mainstream generation methods use frame-by-frame synthesis without adequate temporal consistency constraints, we introduce the Spatiotemporal Attention 3D Network (STA-3D), a novel framework that combines a lightweight spatiotemporal attention module with a 3D convolutional architecture to improve detection robustness. The proposed attention module adopts a symmetric multi-branch architecture, where each branch follows a nearly identical processing pipeline to separately model temporal-channel, temporal-spatial, and intra-spatial correlations. Our framework additionally implements Spatial Pyramid Pooling (SPP) layers along the temporal axis, enabling adaptive modeling regardless of input video length. Furthermore, we mitigate the inherent asymmetry in the quantity of authentic and forged samples by replacing standard cross entropy with focal loss for training. This integration facilitates the simultaneous exploitation of inter-frame temporal discontinuities and intra-frame spatial artifacts, achieving competitive performance across various benchmark datasets under different compression conditions: for the intra-dataset setting on FF++, it improves the average accuracy by 1.09 percentage points compared to existing SOTA, with a more significant gain of 1.63 percentage points under the most challenging C40 compression level (particularly for NeuralTextures, achieving an improvement of 4.05 percentage points); while for the intra-dataset setting, AUC is enhanced by 0.24 percentage points on the DFDC-P dataset. Full article

In this paper, a compressed adaptive image-sensing method based on an overcomplete ridgelet dictionary is proposed. Some low-complexity operations are designed to distinguish between smooth blocks and texture blocks in the compressed domain, and adaptive sampling is performed by assigning different sampling rates to different types of blocks. The efficient, sparse representation of images is achieved by using an overcomplete ridgelet dictionary; at the same time, a reasonable dictionary-partitioning method is designed, which effectively reduces the number of candidate dictionary atoms and greatly improves the speed of classification. Unlike existing methods, the proposed method does not rely on the original signal, and computation is simple, making it particularly suitable for scenarios where a device’s computing power is limited. At the same time, the proposed method can accurately identify smooth image blocks and more reasonably allocate sampling rates to obtain a reconstructed image with better quality. The experimental results show that our method’s image reconstruction quality is superior to that of existing ARCS methods and still maintains low computational complexity. Full article

Two-dimensional human pose estimation (2D HPE) has become a fundamental task in computer vision, driven by growing demands in intelligent surveillance, sports analytics, and healthcare. The rapid advancement of deep learning has led to the development of numerous methods. However, the resulting diversity in research directions and model architectures has made systematic assessment and comparison difficult. This review presents a comprehensive overview of recent advances in 2D HPE, focusing on method classification, technical evolution, and performance evaluation. We classify mainstream approaches by task type (single-person vs. multi-person), output strategy (regression vs. heatmap), and architectural design (top-down vs. bottom-up) and analyze their respective strengths, limitations, and application scenarios. Additionally, we summarize commonly used evaluation metrics and benchmark datasets, such as MPII, COCO, LSP, OCHuman, and CrowdPose. A major contribution of this review is the detailed comparison of the top six models on each benchmark, highlighting their network architectures, input resolutions, evaluation results, and key innovations. In light of current challenges, we also outline future research directions, including model compression, occlusion handling, and cross-domain generalization. This review serves as a valuable reference for researchers seeking both foundational insights and practical guidance in 2D human pose estimation. Full article

High-quality reconstruction of magnetic resonance imaging (MRI) data from undersampled k-space remains a significant challenge in medical imaging. While the integration of compressed sensing and deep learning has notably improved the performance of MRI reconstruction, existing convolutional neural network-based methods are limited by their small receptive fields, which hinders the exploration of global image features. Meanwhile, Swin-Transformer-based approaches struggle with inter-window information interaction and global feature extraction and perform poorly when dealing with complex repetitive structures and similar texture features under undersampling conditions, resulting in suboptimal reconstruction quality. To address these issues, we propose a Symmetry-based Cascaded Dual-Domain Hybrid Attention Network (SCDDHAN). Leveraging the inherent symmetry of medical images, the network combines channel and self-attention to improve global context modeling and local detail restoration. The overlapping window self-attention module is designed with symmetry in mind to improve cross-window information interaction by overlapping adjacent windows and directly linking neighboring regions. This facilitates more accurate detail recovery. The concept of symmetry is deeply embedded in the network design, guiding the model to better capture regular patterns and balanced structures within MRI images. Experimental results demonstrate that under 5× and 10× undersampling conditions, SCDDHAN outperforms existing methods in artifact suppression, achieving more natural edge transitions, clearer complex textures and superior overall performance. This study highlights the potential of integrating symmetry concepts into hybrid attention modules for accelerating MRI reconstruction and offers an efficient, innovative solution for future research in this area. Full article

Ultra-high-performance concrete (UHPC) has gained widespread application across various domains owing to its superior properties. Nevertheless, the high cement content and associated costs present challenges, including significant shrinkage of the cement matrix and economic considerations. Using industrial by-products or waste to replace some raw materials is one of the effective solutions. Meanwhile, China’s ceramic industry generates a large amount of waste every year. Applying ceramics in UHPC can effectively solve these problems. This study explores the use of recycled tile waste as a sustainable alternative to reduce the use of natural aggregates and cement and enhance the performance of UHPC. To investigate the impact of recycled ceramics on the mechanical properties of UHPC, three preparation methods were employed: (1) single incorporation of ceramic tile aggregate (CTA) to replace fine aggregates (0–100%), (2) single incorporation of ceramic tile powder (CTP) to replace cementitious materials (0–20%), and (3) dual incorporation of both CTA and CTP. The effects of different preparation methods and substitution rates on mechanical properties were evaluated through compressive and flexural strength tests, and microstructure analyses were conducted by scanning electron microscopy (SEM) and mercury intrusion porosimetry (MIP). The test results show that the compressive strength and flexural strength of UHPC increased with an increase in the ceramic particle substitution rate and reached the maximum value at a 100% substitution rate. On the contrary, ceramic powder substitution initially reduced the compressive strength, and it slightly recovered at a substitution rate of 10%. However, the bending strength decreased with an increase in the substitution rate of the ceramic powder. When ceramic particles and ceramic powder were used in combination, the compressive strength was the highest when 100% ceramic particles and 20% ceramic powder were used as substitutes. The maximum flexural strength occurred when 100% ceramic particles or 5% ceramic powder was used as a substitute. This study demonstrates that recycled ceramic waste can effectively enhance the mechanical properties of UHPC, providing a sustainable solution for reducing cement consumption and improving the performance of concrete. Full article

Recent advancements in implicit neural representations have shown substantial promise in various domains, particularly in video compression and reconstruction, due to their rapid decoding speed and high adaptability. Building upon the state-of-the-art Neural Representations for Videos, the Expedite Neural Representation for Videos and Hybrid Neural Representation for Videos primarily enhance performance by optimizing and expanding the embedded input of the Neural Representations for Videos network. However, the core module in Neural Representations for Videos network, responsible for video reconstruction, has garnered comparatively less attention. This paper introduces a novel High-frequency Spectrum Hybrid Network, which leverages high-frequency information from the frequency domain to generate detailed image reconstructions. The central component of this approach is the High-frequency Spectrum Hybrid Network block, an innovative extension of the module in Neural Representations for Videos network, which integrates the High-frequency Spectrum Convolution Module into the original framework. The high-frequency spectrum convolution module emphasizes the extraction of high-frequency features through a frequency domain attention mechanism, significantly enhancing both performance and the recovery of local details in video images. As an enhanced module in the Neural Representations for Videos network, it demonstrates exceptional adaptability and versatility, enabling seamless integration into a wide range of existing Neural Representations for Videos network architectures without requiring substantial modifications to achieve improved results. In addition, this work introduces the High-frequency Spectrum loss function and the Multi-scale Feature Reuse Path to further mitigate the issue of blurriness caused by the loss of high-frequency details during image generation. Experimental evaluations confirm that the proposed High-frequency Spectrum Hybrid Network surpasses the performance of the Neural Representations for Videos, the Expedite Neural Representation for Videos, and the Hybrid Neural Representation for Videos, achieving improvements of +5.75 dB, +4.53 dB, and +1.05 dB in peak signal-to-noise ratio, respectively. Full article

Interrupted sampling repeater jamming (ISRJ) is characterized by its coherent processing gains and flexible modulation techniques. ISRJ generates spurious targets along the range, which presents significant challenges to the radar systems. However, existing ISRJ countermeasure methods struggle to eliminate ISRJ signals without compromising the integrity of the real target signal, especially under low-signal-to-noise-ratio (SNR) conditions, resulting in a deteriorated sidelobe and diminished detection performance. We propose a complex-valued encoder–decoder network (CVEDNet) to address these challenges based on signal decomposition. This network offers an end-to-end ISRJ suppression approach, working on complex-valued time-domain signals without the need for additional preprocessing. The encoding and decoding structure suppresses noise components and obtains more compact echo feature representations through layer-by-layer compression and reconstruction. A stacked dual-branch structure and multi-scale dilated convolutions are adopted to further separate the echo signal and ISRJ based on high-dimensional features. A multi-domain combined loss function integrates the waveform and range-pulse-compression information to ensure the amplitude and phase integrity of the reconstructed echo waveform during the training process. The effectiveness of the proposed method was validated in terms of its jamming suppression capability, echo fidelity, and detection performance indicators under low-SNR conditions compared to conventional methods. Full article